Natural weathering, dry air storage, and water curing for long periods of time will have different effects on matrix properties in most fiber reinforced cements. Changes in matrix properties are shown to effect the cracking stress of the composite that will change with time and curing conditions, regardless of changes in fiber properties. The reasons for the differences in the critical fiber volume in uniaxial tension and flexure are explained, and examples are given of 10-year tests on thin cement-based sheets containing networks of fibrillated polypropylene film in which the effects are demonstrated. It is shown that the manufacturer of the composite needs to have an understanding of these problems if the component is to remain ductile for many years in natural weathering conditions.

The Arthur Laing Bridge was constructed in 1975. At a relatively early age of about 6 years it began to suffer damage due to corrosion of the deck reinforcement. A major rehabilitation and resurfacing program was implemented in 1987, which included the installation of a cathodic protection system on about 45 percent of the 21,200 mý deck. This is one of the largest installations of cathodic protection on a reinforced concrete bridge deck. The original deck was milled to a depth of 15 to 25 mm to remove chloride-contaminated concrete. A catalyzed titanium wire mesh anode system was installed on the milled surface after delaminations had been patched. Finally a 50 mm thick low-slump dense concrete overlay was placed. This paper describes the design and construction of the cathodic protection system. Technical details of the cathodic protection and overlay system and construction costs are also presented.

The concrete bridge structural members, called "skew members" (SM), which are positioned from 1.5 m above the sea level to about 20 m down in the sea, and are among the most important elements in bridge construction, were investigated for maintenance purposes after 11 years of service. The underwater arch foundation concrete was also tested. The compressive strength, determined as the average value of 10 concrete cores drilled out from each of two skew members--SM-St. Marko and SM-Mainland--was 62.3 Mpa and 57.4 MPa, respectively. Chlorides had penetrated through the high-alkaline composite by over 20 mm in the splash zone concrete and by over 45 mm in the fully submerged concrete, where {Cl-}-penetration was probably enhanced by hydrostatic pressure. The lack of corrosion of the steel in the concrete, even in the presence of high chloride concentration, could be explained by the absence of oxygen. The gas permeability coefficients Kg determined on the concrete core slices varied in the inner concrete layers of SM-St. Marko from 5.58 to 20.10 x 10-13 cmý and from 0.55 to 2.84 x 10-13 cmý in the concrete at SM-Mainland.

Fiber cement composites developed commercially as replacements for asbestos cement have been under investigation at the Building Research Establishment for a number of years. These include composites containing fibers of glass, polyvinyl alcohol, and cellulose. The suitability of other fibers, including some natural vegetable fibers for cement reinforcement, has also been examined in the past. A summary of the results obtained in some of these studies is presented. Among the composites just mentioned, glass fiber reinforced cement has received the most attention at BRE for historical reasons. In this study, different types of cement--portland, high-alumina, supersulfated, etc.--as well as portland cement modified by pozzolans and polymers--were used. For some of these composites, important mechanical properties after natural weathering for up to 20 years have been determined recently. A review of this work is given, and the results are analyzed in terms of mechanisms that are considered to be important in determining the long-term durability of these materials. It has not been possible yet to devise realistic accelerated aging tests for many of the new fiber cement composites to predict their long-term properties.

As part of an extensive investigation of the durability of roller-compacted concrete pavements, 28 different concrete loads were cast during the summer of 1988. Four types of cement, two different sands (one natural and one manufactured) and two water-cement ratios (0.27 and 0.35) were used to prepare these mixtures, An air-entraining agent was added to half of them. One-third of the test section was moist cured for 7 days, a white curing compound was sprayed on another third and the remaining portion received no special treatment. Samples representative of all mixtures and all curing conditions were taken from the pavement after 28 days. The air-void characteristics of all concretes were determined in accordance with ASTM C 457 and the salt scaling resistance of selected combinations (of the type of mixture and the type of curing) was evaluated using ASTM C 672 on both rolled and sawn surfaces. Results indicate that it is extremely difficult to entrain air in this type of concrete even if fairly large dosages of air-entraining agent are used and mixing time is increased. Despite the lack of spherical air bubbles, good scaling resistances were obtained with the silica fume and the fly ash concretes prepared with the natural sand and cured with a membrane.